Numerical investigation of particle deposition in a triple bifurcation airway due to gravitational sedimentation and inertial impaction

    Numerical investigation of particle deposition in a triple bifurcation airway due to gravitational sedimentation and inertial impaction Xiaole Chen, Yu Feng, Wenqi Zhong, Baobin Sun, Feng Tao PII: DOI: Reference:

S0032-5910(17)30780-5 doi:10.1016/j.powtec.2017.09.050 PTEC 12855

To appear in:

Powder Technology

Received date: Revised date: Accepted date:

21 July 2017 19 September 2017 24 September 2017

Please cite this article as: Xiaole Chen, Yu Feng, Wenqi Zhong, Baobin Sun, Feng Tao, Numerical investigation of particle deposition in a triple bifurcation airway due to gravitational sedimentation and inertial impaction, Powder Technology (2017), doi:10.1016/j.powtec.2017.09.050

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Numerical Investigation of Particle Deposition in a Triple Bifurcation Airway due to Gravitational Sedimentation and Inertial

IP

T

Impaction

a

SC R

Xiaole Chen a, Yu Feng b, Wenqi Zhong a,*, Baobin Sunc, Feng Tao c Key Laboratory of Energy Thermal Conversion and Control of Ministry of Education,

School of Chemical Engineering, Oklahoma State University, Stillwater, OK 74078, USA School of Medicine, Southeast University, Nanjing, Jiangsu Prov., 210096, China

D

c

MA

b

NU

Southeast University, Nanjing, Jiangsu Prov., 210096, China

CE P

Tel: +86 025-83794744

TE

Corresponding author: Dr. Wenqi Zhong

AC

Email: [email protected]

Abstract

Focusing on the particle deposition due to gravitational sedimentation and inertial impaction, the transport and deposition of micron particles with various diameters (i.e., 1 to 7 μm) have been simulated in a triple bifurcation airway at flow rates of 15 to 120 L/min. The airway configurations for generation 11 to generation 14 are determined by matching the Stokes number St and the sedimentation parameter γ (Particuology, 2016, 28:102-113). Air flow distributions, local deposition patterns, and deposition efficiencies are compared between simulations with and without gravity. The results suggest that minimum deposition efficiency exists for large particles 1

ACCEPTED MANUSCRIPT indicating the change of dominant deposition mechanism between sedimentation and impaction. Sedimentation governed particle deposition is also observed for small particles. A revised

IP

T

correlation for predicting DE as a function of Stokes number and sedimentation parameter is

SC R

proposed for 5.78×10-4 < St < 0.226 and 1.54×10-4 < γ < 0.06. These findings could be utilized to design the drug particle property and inhalation waveform for the high-efficiency pulmonary drug

NU

delivery.

Keywords: Inhalable particle; gravitational sedimentation; inertial impaction; deposition

MA

efficiency; airway.

D

1. Introduction

TE

In light of advancing manufacture methods for pharmaceutical particles, pulmonary drug delivery now attracts increasing attentions [1]. To enhance the drug delivery efficiency to local

CE P

lesions in the lung, it is critical to understand the particle transport dynamics and deposition

AC

mechanism in human airways.

For inhaled particulate matters, primary deposition mechanisms include inertial impaction [2-4], gravitational sedimentation [5, 6], as well as Brownian motion. Brownian motion mainly affects submicron particles [7-9]. In contrast, for micron particles, inertial impaction governs the deposition in the upper airway, while sedimentation plays the dominant role in the alveolar region. Compared to the many research investigating the different parameters that affect the two different deposition mechanisms [10-27], there are limited investigations regarding the transition of the dominant deposition mechanism [5, 28-31]. Specifically, for a given particle diameter and inhalation flow rate, the key question remains unanswered, e.g., is it possible to find an airway

2

ACCEPTED MANUSCRIPT generation in which inertial impaction and sedimentation provide a similar contribution to the particle deposition (see Fig. 1)? Such investigations will contribute to the discovery of the airway

IP

T

region with minimum deposition efficiency (DE) associated with specific particle properties and

SC R

breathing patterns. Therefore, studying the contributions of different deposition mechanisms could be utilized for optimization of drug formulation and pulmonary drug delivery operations. For example, drug particle properties and breathing pattern could be designed on a patient-specific

NU

level, to match these conditions with the lowest regional DEs in the transition regions, thereby

MA

guarantee a higher drug delivery efficiency in the alveolar region for a better translocation performance into systemic regions via blood circulation [32]. Another potential application

D

example is the pulmonary targeted drug delivery to treat localized lung tumor [33] or diseased

TE

areas [34] in the transition regions by controlling the operational parameters for drug inhalation,

CE P

such as inhalation flow rate and breath holding duration. In such a method, the local deposition will be increased by the enhanced sedimentation effect.

AC

Hofmann et al. simulated the simulated the transport of 10μm particles from G15 ("G" stands for generation) to G16 with gravity [35]. Ma et al. [36] compared the cross-sectional velocity contours and particle trajectories between CFD simulations and experimental results in a scaled-up alveolated airway model. The comparison indicates that CFD techniques can provide an accurate prediction for air flow and particle transport in alveoli. Furthermore, Ma and Darquenne studied the depositions of 1 μm and 3 μm particles in models of the human alveolar sac and terminal acinar bifurcation under rhythmic wall motion for two breathing conditions [37]. Sznitman et al. investigated the transport and deposition of 1 μm and 3 μm particles in two acinar models, i.e., a simple alveolar duct and a space-filling acinar branching tree, with sinusoidal breathing waveform 3

ACCEPTED MANUSCRIPT [30]. The results suggested that the dominant deposition mechanism changes from convection to sedimentation within a span of 2 μm. Piglione et al. concluded that Froude number is the key

IP

T

parameter to determine the relative importance of impaction and settling [38]. However, Fr does

SC R

not include the length parameters of the airway, i.e., length and diameter of the airway duct. Kleinstreuer et al. proposed the non-dimensional sedimentation parameter γ to indicate the significance of sedimentation when simulating the particle transport and deposition in a G6-G9

NU

airway model [5]. Their results claimed that gravitational deposition might become dominant for

MA

large particles (> 5 μm) at a flow rate of 3.75 L/min. Our previous experimental measurements of the deposition of fibers indicated that the deposition fraction (DF) increases linearly with the

D

increasing γ when the fibers are relatively large [39]. Additionally, a method matching Stokes

TE

number St and sedimentation parameter γ were proposed to correlate similar deposition

CE P

characteristics in the lower airway with the experiment results. Still, existing papers do not provide systematic investigations and conclusions on the relationship between dominant

AC

deposition mechanism types and different airway generations.

To fill the knowledge gap and provide an insightful discussion, this paper focuses on simulating the particle deposition due to sedimentation and inertial impaction, the transport and deposition of micron particles with various diameters (i.e., 1 to 7 μm) in a G11-G14 airway at flow rates of 15 to 120 L/min using Computational Fluid-Particle Dynamics (CFPD) method. Extra simulations without gravity are also performed for comparison purposes. Airflow distributions, deposition patterns, and deposition efficiencies are compared. The minimum DEs are found for large particles, and gravitational sedimentation governed deposition is also observed for small particles. A revised correlation of DE prediction is proposed, which is a function of Stokes number 4

ACCEPTED MANUSCRIPT and sedimentation parameter.

T

2. Triple Bifurcation Airway Unit

IP

To investigate the particle deposition as they travel into the deeper airways, it is essential to

SC R

determine the critical generation where gravitational sedimentation becomes dominant compared to inertial impaction. Therefore, sedimentation parameter γ is introduced which is defined as [5]:  vsettling   L     cos  U  D

(1)

NU

 

MA

where vsettling is the particle settling velocity, U is the average air velocity, L is the length of the airway, D is the diameter of the airway,  is the inclination angle measured relative to the

D

horizontal. The sedimentation parameter includes two more length parameters of the airway, i.e., L

TE

and D, than the Froude number Fr  U / gD . According to this advantage, Kleinstreuer et al. [5]

airways.

CE P

suggested that  is more suitable than Fr for the estimation of particle sedimentation in lower

AC

Our previous experiments indicated that the effect of the gravity on particle deposition could not be neglected when 0.0228 < γ < 0.247 [39]. The results also suggested that the experimental data could be used to determine the appropriate inhalation patterns leading to the dominant sedimentation effect in a designated airway generation, by matching the Stokes number St and sedimentation parameter γ. Specifically, St is defined as

St 

 p d p2U 18 D

(2)

where  p is the particle density, d p is the particle diameter and μ are the dynamic viscosity of air. 5

ACCEPTED MANUSCRIPT Specifically, once particle properties are known, and the airway generation is assigned, the average air velocity U, i.e., inhalation flow rate at the oral/nasal inlets, could be calculated from

IP

T

Eqs. (1) and (2) using γ and St obtained from experiments [39]. Note that it is possible that the

SC R

calculated flow rate is not within the reasonable range (i.e., 15 to 120 L/min). However, for those generations with reasonable flow rates, the particle deposition would share similar characteristics

NU

as the experiments, i.e., governed by gravitational sedimentation.

Concluded by Chen et al. [39], the pulmonary generation G10 to G14 cover a relatively wide

MA

range of reasonable inhalation flow rates for both St and γ. Simulation results suggested that four generations of the pulmonary airway could accurately predict the effect of the air flow of the

TE

D

upper and lower generations on the particle deposition [40, 41]. Therefore, a triple bifurcation airway model including G11 to G14 is chosen for the following analysis of the air flow

CE P

distributions and particle depositions.

AC

3. Methodology

3.1 Governing Equations for Airflow For inhalation flow rate 15
u  0

(u  )u 

u p     [ (u  (u )tr )] t 

6

(3)

(4)

ACCEPTED MANUSCRIPT where u is the air velocity, t is time, p is the air pressure, ρ is the air density, and (u )tr is the

T

transpose of u .

IP

3.2 Governing Equations for Particle Transport

SC R

Assuming the particle suspension in the lower airways is dilute, the particle-particle interaction are neglected. Additionally, Brownian motion effect is negligible. Thus, the effects of

du p dt

 FD  m p (  p   )

MA

mp

NU

drag force and gravity are considered for predicting the particle trajectory, i.e.,

g

p

(5)

where FD is the drag force, which is defined as

TE

D

1 FD   d p2CD  u  u p  u  u p 8

(6)

CE P

where CD is the drag coefficient, which is given as [42]: CD 

a1 a  2  a3 Re p Re2p

(7)

AC

where a1 , a2 , and a3 are coefficients. The particle Reynolds number is defined as Re p 

 |u  u p|d p 

(8)

3.3 Geometry and mesh Due to the limitation of the current technology, it is still difficult to obtain the realistic airway model directly from the CT or MRI scan. Thus, a triple bifurcation airway unit from G11 to G14 was constructed based on Weibel's 23-generation pulmonary model [43] (see Fig. 2). The bifurcation angle for each generation is 60°. Inner and lateral curves with different diameters were used to generate smooth connections between airway generations. The details of the geometric parameters are summarized in Table 1. 7

ACCEPTED MANUSCRIPT Tetrahedral mesh with prism layers was generated for the numerical simulation. To obtain an airflow independent of the mesh number, four meshes with different mesh number, i.e., 420,693,

IP

T

1,076,796, 1,939,271 and 2,738,750, were generated for the mesh independence test. The

SC R

dimensionless velocity profiles of the mid-plane in G12, G13.1, and G13.2 were compared at the flow rate of 120 L/min (see Fig. 3). Note that the average velocity at the inlet of G11 is calculated based on the assumption that each airway generation would have two daughter tubes. Thus, the

NU

actual volume flow rate in G11 is 120 / 2^11 L/min, i.e., 0.0586 L/min, if the inhalation flow rate

MA

is 120 L/min at the oral / nasal inlet. Considering the differences of dimensionless velocity profiles between the 1.93 million and 2.73 million meshes in Fig. 3 are all smaller than 5%, the mesh with

TE

3.4 Numerical Setup

D

1.93 mesh cells was used as the final mesh for this study.

CE P

In light of the dilute particle suspension, one-way coupling was applied in the simulations. The airflow distributions were solved first for different inhalation flow rates at the inlet of

AC

oral/nasal cavity, i.e., 15, 30, 45, 60, 90 and 120 L/min. User defined function (UDF) was used to generate the parabolic velocity inlet for each case [44]. The simulation for the airflow was considered converged when the residual became less than 10-5. To obtain the correlation between particle deposition efficiency and Stokes number in a relatively wide range, the simulations of particle transport and deposition were carried out for particles with a density of 1550 kg / m3 and diameters from 1 to 7 μm. An in-house C++ code was generated to determine the initial position of the particle at the inlet of G11. It generates a random distribution with a parabolic probability density which is proportional to the air velocity. The particle is assumed to have the same velocity as the local air at the inlet. For each case, 50,000 8

ACCEPTED MANUSCRIPT particles were generated, and their trajectories were calculated respectively. Particle deposition was assumed to occur when the particle contacted with the airway boundary. The locations of

IP

T

deposited and escaped particles were exported via another UDF when the deposition or the escape

SC R

happened. The Stokes numbers and sedimentation parameters for the different cases used in this study are presented in Table 2 and Table 3, respectively. The gravity direction was vertical to the mid-plane towards the –x direction as shown in Fig. 2. An extra set of simulations without gravity

NU

were also performed to disseminate the deposition caused by inertial impaction from gravitational

4. Results and Discussion

D

4.1 Airflow distributions

MA

sedimentation.

TE

Fig. 4 illustrates the air velocity contours of the mid-plane and the cross-sectional velocity

CE P

contours and secondary flow vectors at the flow rate of 15, 60 and 120 L/min. The reference vectors are adjusted to present the secondary flows. It is evident that the velocity distributions of

AC

the air in Fig. 4 show the distinct difference from that of the upper lung generations. The simulations in the G3-G6 airway by Zhang et al. showed that the velocity profiles in the daughter tubes were skewed indicating strong secondary flows, e.g., the maximum secondary velocity could reach 21% of the mean inlet velocity [45]. In contrast, the spatial positions of the highest velocity in the cross sections (i.e., the E-E', F-F' and G'-G cross sections are shown in Fig. 4) locate in the center of the airway tubes in this study. Thus, the secondary flow is relatively weak, i.e., 2 to 3 magnitudes smaller than maximum velocity in the cross section. The flow distributions at the outlets (i.e., G14.1 to G14.2) are almost ideally even(see Table 4).

9

ACCEPTED MANUSCRIPT 4.2 Final particle distributions The final distributions of the 2.5 μm particles, i.e., the locations of deposited and escaped

IP

T

particles, are shown in Fig. 5 for the inhalation flow rates of 15, 60 and 120 L/min conditions with

SC R

and without gravity. Considering the airway model is symmetric about plane y = 0, positions of the escaped particles at the G14.5 to G14.8 outlets are mirrored to G14.4 to G14.1 outlets shown in Fig. 5.

NU

For the simulation cases without gravity, the main deposition mechanism would be inertial

MA

impaction. As shown in Figs. 5 (a), (c) and (e), this type of deposition enhances with the increase of flow rate. Despite few scattered deposition on the airway tubes due to the secondary flow, most

D

deposition occurred in the bifurcation regions. It is interesting to find that the distributions of

TE

escaped particles at the outlets are similar to the parabolic-shaped probability density distribution

CE P

when releasing the particles at the inlet of G11. When the effect of the gravity on the particle is considered in the simulations, the deposition

AC

patterns (see Figs. 5(b), (d) and (f)) show a significant difference compared to the ones without gravity. Almost all particle deposition occurred on the boundary of the bottom half of the airway except the depositions near the bifurcating regions. This indicates that gravitational sedimentation is the dominant deposition mechanism. This significant difference is caused by the slow airflow in the G11 to G14 airway. When the effect of the gravity is not included, the velocities of the particles are relatively small due to the slow airflow, and therefore, most particles can well follow the airflow in the bifurcating region instead of colliding onto the airway boundary. In contrast, the particles gradually settle down and deposit on the bottom of the airway when gravity is considered. It is also worth noting that the number of deposited particles reduces with increased flow rate, 10

ACCEPTED MANUSCRIPT which suggests that the particles are entrained to the subsequent airways instead of depositing on the bottom of the airway. There are crescent areas free of particles at the four outlets. This

IP

T

indicates that the particles move towards the direction of the gravity while the secondary flow is

SC R

too weak to bring the particle upward due to Dean's flow.

4.3 Deposition Efficiencies 4.3.1 DE vs. St

NU

The particle deposition efficiencies obtained from the simulations with and without gravity

MA

are plotted against the Stokes number as shown in Fig. 6. The DEs for simulations under 6 different flow rate conditions without gravity, which are shown as black dots, are concentrated. data

fit



the



correlation

proposed

by

Kim

and

Fisher

[10],

i.e.,

D

These

TE

DE  1  1 /  C1St C2  1  100% , very well (see the black dash line in Fig. 6). If C1 = 14.76 and

CE P

C2 = 2.34, for variance analysis, the adjusted R2 value would be 0.989.

However, the DEs for simulations with gravity are more scattered in Fig. 6. Thus, these data

AC

points are organized by two individual parameters, i.e., the same flow rate and the same particle diameter. For the same flow rate (solid black lines in Fig. 6), the DEs all increase with the growth of St. The increasing trend is similar to Kim and Fisher's correlation [10]. Still, the DEs under different inhalation flow rate conditions have significant divergence for the same St. It suggests that there is an extra factor, gravity, in this case, affecting the DE.

For the same particle diameter (solid colorful lines in Fig. 6), the particle deposition behaviors are different from the results of the upper airway. It is well known that the DEs increase with higher flow rate from the oral [46] or nasal cavity [47, 48] to the upper lung generations [4, 49]. The DE-curves of 1 μm and 2.5 μm decrease with the increase of St. This is in contradiction 11

ACCEPTED MANUSCRIPT to the traditional suggestion for pulmonary drug delivery, which encourages slow inhalation for drug administration. Thus, a careful design of the inhalation waveform might be considered for the

IP

T

inhalable drug delivery aiming the deeper lung regions. Meanwhile, the curves of 5 μm and 7 μm

SC R

decrease first, reach the lowest DE points, and then increase with St again. It proves the assumption that when the effect of gravity on the particle deposition increases and air velocity gradually decreases towards the lower airway generations, a minimum DE point exists. Therefore,

NU

the DE cannot be correlated by St alone, and the effect of gravity should be included.

MA

4.3.2 DE vs. γ

In the previous experiment, the DE increases linearly with the increase of γ under low

TE

D

inhalation flow rate conditions [39]. For the same flow rate, the DE does increase linearly with increasing γ (see Fig. 7(a)). However, slopes still vary between different flow rates.

CE P

Assuming that the depositions caused by inertial impaction and sedimentation are independent to each other, the DE caused by sedimentation could be obtained by subtracting the

AC

DE of the simulations without gravity from the DE of simulations with gravity. Fig. 7(b) plots these DEs from the direct subtraction against γ. The linear correlation for these DEs excluding inertial impaction is relatively accurate. However, the DEs under the flow rate conditions in the middle range, i.e., 45 and 60 L/min, still have minor deviations from the linear correlation, which suggests that the depositions caused by inertial impaction and sedimentation are related.

4.3.3 Correlation for DE Kleinstreuer et al. proposed a correlation for particle deposition due to sedimentation and inertial impaction in a G6-G9 airway [5], i.e.,

12

ACCEPTED MANUSCRIPT DE   0.974St1.26  0.4558 0.775  13.655St  100%

(9)

From the aspect of deposition mechanism, the inertial impaction of the particle is caused by

IP

T

the entrainment of the air. It is reasonable to correlate the DE with the term St n , i.e., the U n .

SC R

However, the deposition caused by particle sedimentation correlates to the sedimentation parameter  linearly based on the previous discussion and the experiment [39]. Also considering the possible deposition due to the combined effects of sedimentation and impaction, the third term

NU

n should be St  . Thus, the deposition efficiency in the G11 to G14 airway can be correlated as

(10)

MA

DE   aSt b  c  dSt b   100%

For the simulations of this study (5.78×10-4 < St < 0.226, 1.54×10-4 < γ < 0.06), a = 3.397, b =

TE

D

1.519, c = 10.885, and d = -1.284 yield the best fit (see Fig. 8), where adjusted R2 = 0.996. This result is improved compared to the value of 0.93 obtained by Kleinstreuer et al. [5].

CE P

5. Conclusions

Particle deposition due to gravitational sedimentation and inertial impaction in a triple

AC

bifurcation airway have been analyzed for particle diameter ranging from 1 μm to 7 μm and inhalation flow rate between 15 and 120 L/min. The airway generations, i.e., G11 to G14, was selected according to the approach of matching Stokes number and sedimentation parameter, which was proposed based on the observation of our previous experimental measurements [39]. The particle deposition patterns and regional deposition efficiencies of simulations with and without gravity have been compared. The following conclusions can be drawn:

(1) In contrast to the upper lung generations, the airflow distributions in the lower generations are almost equal in each daughter tubes.

13

ACCEPTED MANUSCRIPT (2) There are distinct differences in the deposition patterns of 2.5 μm particles under different inhalation flow rates conditions when comparing the simulation results with and without gravity.

IP

T

Particle deposition dominated by gravitational sedimentation is observed.

SC R

(3) Particle deposition efficiencies have been analyzed as functions of Stokes number and sedimentation parameter individually. Minimum deposition efficiency is existed for large particles,

NU

i.e., 5 μm and 7 μm in diameters when considering gravity. The deposition efficiency decreases with increased Stokes number for small particles, which indicates that the dominant deposition

MA

mechanism is sedimentation.

D

(4) A new correlation function is proposed for predicting particle deposition efficiency when

TE

5.78×10-4 < St < 0.226 and 1.54×10-4 < γ < 0.06, i.e.,

(11)

CE P

DE   3.397St1.519  10.885  1.284St1.519  100%

These deposition characteristics could be utilized for patient-specific pulmonary targeted drug

AC

delivery in the lower airway or the alveolar regions. For example, breath-holding or slow inhalation should be adopted for drug targeting the lower lung generations (G11 to G14 in this study) when the drug particle is traveling through these generations. Higher flow rate at the end of inhalation could decrease the deposition due to sedimentation in the lower airway, if the drug particle is designated to deposit in alveolar regions, e.g., the inhaled insulin. This study is limited to the symmetric-in-plane model for G11 to G14 airway. Future works may focus on the combined effects of gravitational sedimentation and inertial impaction in airway units in wider range of airway generations with asymmetric geometries and out-of-plane bifurcations as well as different inclination angles to improve the correlation function.

14

ACCEPTED MANUSCRIPT Furthermore, the effects of the sedimentation and impaction on macroscopic scale could be evaluated, if the improved correlation function can be used with statistics of the airway number,

IP

T

size and inclination angle for different generations.

SC R

Acknowledgements

The authors gratefully acknowledge the financial support of the National Natural Science

NU

Foundation of China (grant No. 51606041), National Natural Science Foundation of Jiangsu Prov. (grant No. BK20160688). The Fundamental Research Funds for the Central Universities are also

MA

acknowledged.

D

References

TE

[1] C. Kleinstreuer, Z. Zhang, J. Donohue, Targeted drug-aerosol delivery in the human respiratory system, Annual Review of Biomedical Engineering, 10 (2008) 195-220.

CE P

[2] Y. Zhang, T.L. Chia, W.H. Finlay, Experimental measurement and numerical study of particle deposition in highly idealized mouth-throat models, Aerosol Science and Technology, 40 (2006)

AC

361-372.

[3] Y. Zhang, W.H. Finlay, E.A. Matida, Particle deposition measurements and numerical simulation in a highly idealized mouth–throat, Journal of Aerosol Science, 35 (2004) 789-803. [4] Z. Zhang, C. Kleinstreuer, C. Kim, Gas–solid two-phase flow in a triple bifurcation lung airway model, International Journal of Multiphase Flow, 28 (2002) 1021-1046. [5] C. Kleinstreuer, Z. Zhang, C.S. Kim, Combined inertial and gravitational deposition of microparticles in small model airways of a human respiratory system, Journal of Aerosol Science, 38 (2007) 1047-1061. [6] S. Chhabra, A.K. Prasad, Flow and Particle Dispersion in a Pulmonary Alveolus—Part II: 15

ACCEPTED MANUSCRIPT Effect of Gravity on Particle Transport, Journal of Biomechanical Engineering, 132 (2010) 051010.

IP

T

[7] Y.-S. Cheng, Y. Yamada, H.-C. Yeh, D.L. Swift, Diffusional deposition of ultrafine aerosols in

SC R

a human nasal cast, Journal of Aerosol Science, 19 (1988) 741-751.

[8] P. Zamankhan, G. Ahmadi, Z. Wang, P.K. Hopke, Y.-S. Cheng, W.C. Su, D. Leonard, Airflow and deposition of nano-particles in a human nasal cavity, Aerosol Science and Technology, 40

NU

(2006) 463-476.

MA

[9] P.F. Ghalati, E. Keshavarzian, O. Abouali, A. Faramarzi, J. Tu, A. Shakibafard, Numerical analysis of micro-and nano-particle deposition in a realistic human upper airway, Computers in

D

Biology and Medicine, 42 (2012) 39-49.

TE

[10] C.S. Kim, D.M. Fisher, Deposition characteristics of aerosol particles in sequentially

CE P

bifurcating airway models, Aerosol Science & Technology, 31 (1999) 198-220. [11] C.S. Kim, D.M. Fisher, D.J. Lutz, T.R. Gerrity, Particle deposition in bifurcating airway

AC

models with varying airway geometry, Journal of Aerosol Science, 25 (1994) 567-581. [12] Y. Zhou, Y.-S. Cheng, Particle deposition in a cast of human tracheobronchial airways, Aerosol Science and Technology, 39 (2005) 492-500. [13] T. Sera, K. Uesugi, N. Yagi, H. Yokota, Numerical simulation of airflow and microparticle deposition in a synchrotron micro-CT-based pulmonary acinus model, Computer Methods in Biomechanics and Biomedical Engineering, 18 (2015) 1427-1435. [14] Y.S. Cheng, Y. Zhou, W.-C. Su, Deposition of Particles in Human Mouth–Throat Replicas and a USP Induction Port, Journal of Aerosol Medicine and Pulmonary Drug Delivery, 27 (2014) 1-9. 16

ACCEPTED MANUSCRIPT [15] K. Inthavong, J. Tu, C. Heschl, Micron particle deposition in the nasal cavity using the v2–f model, Computers & Fluids, 51 (2011) 184-188.

IP

T

[16] Z. Zhang, C. Kleinstreuer, C.S. Kim, Effects of curved inlet tubes on air flow and particle

SC R

deposition in bifurcating lung models, Journal of Biomechanics, 34 (2001) 659-669. [17] I. Balásházy, T.B. Martonen, W. Hofmann, Simultaneous sedimentation and impaction of aerosols in two-dimensional channel bends, Aerosol Science and Technology, 13 (1990) 20-34.

NU

[18] T.R. Gerrity, P.S. Lee, F.J. Hass, A. Marinelli, P. Werner, R. Lourenco, Calculated deposition

MA

of inhaled particles in the airway generations of normal subjects, Journal of Applied Physiology, 47 (1979) 867-873.

D

[19] L. Augusto, J. Gonçalves, G. Lopes, CFD Evaluation of the Influence of Physical

TE

Mechanisms, Particle Size, and Breathing Condition on the Deposition of Particulates in a Triple

CE P

Bifurcation Airway, Water, Air, & Soil Pollution, 227 (2016) 56. [20] L. Augusto, G. Lopes, J. Gonçalves, A CFD study of deposition of pharmaceutical aerosols

AC

under different respiratory conditions, Brazilian Journal of Chemical Engineering, 33 (2016) 549-558.

[21] Y. Liu, E.A. Matida, M.R. Johnson, Experimental measurements and computational modeling of aerosol deposition in the Carleton-Civic standardized human nasal cavity, Journal of Aerosol Science, 41 (2010) 569-586. [22] P.W. Longest, S. Vinchurkar, Effects of mesh style and grid convergence on particle deposition in bifurcating airway models with comparisons to experimental data, Medical Engineering & Physics, 29 (2007) 350-366. [23] G. Tian, P.W. Longest, G. Su, R.L. Walenga, M. Hindle, Development of a stochastic 17

ACCEPTED MANUSCRIPT individual path (SIP) model for predicting the tracheobronchial deposition of pharmaceutical aerosols: effects of transient inhalation and sampling the airways, Journal of Aerosol Science, 42

IP

T

(2011) 781-799.

SC R

[24] X. Chen, W. Zhong, B. Sun, B. Jin, X. Zhou, Study on gas/solid flow in an obstructed pulmonary airway with transient flow based on CFD–DPM approach, Powder Technology, 217 (2012) 252-260.

NU

[25] X. Chen, W. Zhong, X. Zhou, B. Jin, B. Sun, CFD–DEM simulation of particle transport and

MA

deposition in pulmonary airway, Powder Technology, 228 (2012) 309-318. [26] L. Golshahi, G. Tian, M. Azimi, Y.-J. Son, R. Walenga, P.W. Longest, M. Hindle, The use of

D

condensational growth methods for efficient drug delivery to the lungs during noninvasive

TE

ventilation high flow therapy, Pharmaceutical Research, 30 (2013) 2917-2930.

CE P

[27] Á. Farkas, I. Balásházy, Quantification of particle deposition in asymmetrical tracheobronchial model geometry, Computers in Biology and Medicine, 38 (2008) 508-518.

AC

[28] B. Asgharian, S. Anjilvel, Inertial and gravitational deposition of particles in a square cross section bifurcating airway, Aerosol Science and Technology, 20 (1994) 177-193. [29] I. Balásházy, T.B. Martonen, W. Hofmann, Inertial impaction and gravitational deposition of aerosols in curved tubes and airway bifurcations, Aerosol Science and Technology, 13 (1990) 308-321. [30] J. Sznitman, T. Heimsch, J.H. Wildhaber, A. Tsuda, T. RÃļsgen, Respiratory flow phenomena and gravitational deposition in a three-dimensional space-filling model of the pulmonary acinar tree, Journal of Biomechanical Engineering, 131 (2009) 031010. [31] B. Asgharian, W. Hofmann, R. Bergmann, Particle deposition in a multiple-path model of the 18

ACCEPTED MANUSCRIPT human lung, Aerosol Science & Technology, 34 (2001) 332-339. [32] C. Kleinstreuer, Z. Zhang, Optimal drug-aerosol delivery to predetermined lung sites, Journal

IP

T

of Heat Transfer, 133 (2011) 011002.

SC R

[33] C. Kleinstreuer, Z. Zhang, Targeted Drug Aeroso Deposition Analysis for a Four-Generation Lung Airway Model With Hemispherical Tumors, Journal of Biomechanical Engineering, 125 (2003) 197-206.

NU

[34] H. Luo, Y. Liu, X. Yang, Particle deposition in obstructed airways, Journal of Biomechanics,

MA

40 (2007) 3096-3104.

[35] W. Hofmann, I. Balásházy, L. Koblinger, The effect of gravity on particle deposition patterns

D

in bronchial airway bifurcations, Journal of Aerosol Science, 26 (1995) 1161-1168.

TE

[36] B. Ma, V. Ruwet, P. Corieri, R. Theunissen, M. Riethmuller, C. Darquenne, CFD simulation

CE P

and experimental validation of fluid flow and particle transport in a model of alveolated airways, Journal of Aerosol Science, 40 (2009) 403-414.

AC

[37] B. Ma, C. Darquenne, Aerosol deposition characteristics in distal acinar airways under cyclic breathing conditions, Journal of Applied Physiology, 110 (2011) 1271-1282. [38] M.C. Piglione, D. Fontana, M. Vanni, Simulation of particle deposition in human central airways, European Journal of Mechanics-B/Fluids, 31 (2012) 91-101. [39] X. Chen, W. Zhong, J. Tom, C. Kleinstreuer, Y. Feng, X. He, Experimental-computational study of fibrous particle transport and deposition in a bifurcating lung model, Particuology, 28 (2016) 102-113. [40] Z. Zhang, C. Kleinstreuer, C. Kim, Flow structure and particle transport in a triple bifurcation airway model, Journal of Fluids Engineering, 123 (2001) 320-330. 19

ACCEPTED MANUSCRIPT [41] Z. Zhang, C. Kleinstreuer, C.S. Kim, Comparison of analytical and CFD models with regard to micron particle deposition in a human 16-generation tracheobronchial airway model, Journal of

IP

T

Aerosol Science, 40 (2009) 16-28.

SC R

[42] S. Morsi, A. Alexander, An investigation of particle trajectories in two-phase flow systems, Journal of Fluid Mechanics, 55 (1972) 193-208.

[43] E.R. Weibel, Morphometry of the human lung, Academic Press, New York, 1963.

NU

[44] Fluent 14.5 users guide, ANSYS Inc., Canonsburg, PA, USA, 2012.

MA

[45] Z. Zhang, C. Kleinstreuer, C. Kim, Cyclic micron-size particle inhalation and deposition in a triple bifurcation lung airway model, Journal of Aerosol Science, 33 (2002) 257-281.

D

[46] B. Grgic, W. Finlay, A. Heenan, Regional aerosol deposition and flow measurements in an

TE

idealized mouth and throat, Journal of Aerosol Science, 35 (2004) 21-32.

CE P

[47] Y. Liu, E.A. Matida, J. Gu, M.R. Johnson, Numerical simulation of aerosol deposition in a 3-D human nasal cavity using RANS, RANS/EIM, and LES, Journal of Aerosol Science, 38 (2007)

AC

683-700.

[48] X. Li, K. Inthavong, J. Tu, Particle inhalation and deposition in a human nasal cavity from the external surrounding environment, Building and Environment, 47 (2012) 32-39. [49] K. Inthavong, J. Wen, Z. Tian, J. Tu, Numerical study of fibre deposition in a human nasal cavity, Journal of Aerosol Science, 39 (2008) 253-265.

20

ACCEPTED MANUSCRIPT Table Captions

IP

T

Table 1 Geometric parameters for the G11-G14 airway model

SC R

Table 2. Stokes numbers for particles with different diameters under different flow rate conditions

NU

Table 3. Sedimentation parameters for particles with different diameter under different flow rate

MA

conditions

AC

CE P

TE

D

Table 4. Flow distributions at the outlets under different flow rate conditions

21

ACCEPTED MANUSCRIPT Figure Captions

SC R

Figure 2. Schematics of the G11 to G14 airway model and mesh

IP

T

Figure 1. Schematics of particle deposition mechanisms in human airway

NU

Figure 3. Comparisons of velocity profiles in different generations for mesh independence test: (a)

MA

G12; (b) G13.1; (c) G14.1

TE

(b) 60 L/min; (c) 120 L/min

D

Figure 4. Airflow distributions in the triple bifurcation airway at different flow rates: (a) 15 L/min;

CE P

Figure 5. Comparison of deposition patterns and locations of escaped particles between simulations with and without gravity: (a) 15 L/min, without gravity; (b) 15 L/min, with gravity; (c)

AC

60 L/min, without gravity; (d) 60 L/min, with gravity; (e) 120 L/min, without gravity; (f) 120 L/min, with gravity

Figure 6. DE vs. St for simulations in the triple bifurcation airway with and without gravity

Figure 7. DE vs. γ for simulations in the triple bifurcation airway with gravity: (a) DE vs. γ; (b) DE excluding inertial impaction vs. γ, and linear correlation

22

ACCEPTED MANUSCRIPT Figure 8. Prediction of deposition efficiency in the triple bifurcation airway with gravity using the

AC

CE P

TE

D

MA

NU

SC R

IP

T

new correlation

23

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

D

MA

Figure 1

24

AC

Figure 2

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

25

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

Figure 3a

26

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

Figure 3b

27

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

Figure 3c

28

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Figure 4a

29

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Figure 4b

30

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Figure 4c

31

AC

Figure 5a

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

32

AC

Figure 5b

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

33

AC

Figure 5c

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

34

AC

Figure 5d

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

35

AC

Figure 5e

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

36

AC

Figure 5f

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

37

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

Figure 6

38

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

Figure 7a

39

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

Figure 7b

40

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

TE

Figure 8

41

ACCEPTED MANUSCRIPT Table 1 Geometric parameters for the G11-G14 airway model Geometrical size（mm） Diameter, Di

Length, Li

Lateral radius, Ri

Inner radius, ri

G11 G12 G13 G14

1.09 0.95 0.82 0.74

3.213 2.268 2.165 2.300

2.565 3.854 1.998 -

0.095 0.082 0.074 -

IP

SC R NU MA D TE CE P AC

42

T

Airway Generation

ACCEPTED MANUSCRIPT Table 2. Stokes numbers for particles with different diameters under different flow rate conditions

2.31E-3 4.62E-3 6.93E-3 9.24E-3 1.39E-2 1.85E-2

3.61E-3 7.22E-3 1.08E-2 1.44E-2 2.17E-2 2.89E-2

5.20E-3 1.04E-2 1.56E-2 2.08E-2 3.12E-2 4.16E-2

MA D TE CE P 43

9.24E-3 1.85E-2 2.77E-2 3.70E-2 5.55E-2 7.39E-2

5

7

1.44E-2 2.89E-2 4.33E-2 5.78E-2 8.66E-2 1.16E-1

2.83E-2 5.66E-2 8.49E-2 1.13E-1 1.70E-1 2.26E-1

T

5.78E-4 1.16E-3 1.73E-3 2.31E-3 3.47E-3 4.62E-3

IP

2

NU

9.76 19.52 29.28 39.05 58.57 78.09

1

AC

15 30 45 60 90 120

Particle diameter (μm) 2.5 3 4

SC R

Flow rate Reynolds (L/min) Number

ACCEPTED MANUSCRIPT Table 3. Sedimentation parameters for particles with different diameter under different flow rate conditions

7.70E-3 3.85E-3 2.57E-3 1.93E-3 1.28E-3 9.63E-4

1.11E-2 5.55E-3 3.70E-3 2.77E-3 1.85E-3 1.39E-3

MA D TE CE P 44

1.97E-2 9.86E-3 6.57E-3 4.93E-3 3.29E-3 2.47E-3

5

7

3.08E-2 1.54E-2 1.03E-2 7.70E-3 5.14E-3 3.85E-3

6.04E-2 3.02E-2 2.01E-2 1.51E-2 1.01E-2 7.55E-3

T

4.93E-3 2.47E-3 1.64E-3 1.23E-3 8.22E-4 6.16E-4

IP

1.23E-3 6.16E-4 4.11E-4 3.08E-4 2.05E-4 1.54E-4

NU

9.76 19.52 29.28 39.05 58.57 78.09

2

AC

15 30 45 60 90 120

Particle diameter (μm) 2.5 3 4

1

SC R

Flow rate Reynolds (L/min) Number

ACCEPTED MANUSCRIPT Table 4. Flow distributions at the outlets under different flow rate conditions G14.1

G14.2

12.5090 12.5064 12.5088

12.4889 12.4911 12.4975

12.5036 12.5046 12.4999

AC

CE P

TE

D

MA

NU

SC R

IP

15 60 120

Ratio (%) G14.3

T

Inhalation flow rate (L/s)

45

G14.4

12.4986 12.4979 12.4939

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Graphical abstract

46

ACCEPTED MANUSCRIPT Highlights

CE P

TE

D

MA

NU

SC R

IP

T

Particle transport and deposition have been simulated in a G11-G14 human airway Simulations focused on the particle deposition due to sedimentation and impaction. Minimum deposition efficiency exists for large particles. Sedimentation governed particle deposition is observed for small particles. A revised correlation for predicting DE is proposed.

AC

    

47